POWN: Prototypical Open-World Node Classification

Marcel Hoffmann, Lukas Galke, Ansgar Scherp
Proceedings of The 3rd Conference on Lifelong Learning Agents, PMLR 274:672-691, 2025.

Abstract

We consider the problem of true open-world semi-supervised node classification, in which nodes in a graph either belong to known or new classes, with the latter not present during training. Existing methods detect and reject new classes but fail to distinguish between different new classes. We adapt existing methods and show they do not solve the problem sufficiently. We introduce a novel end-to-end approach for classification into known classes and new classes based on class prototypes, which we call Prototypical Open-World Learning for Node Classification (POWN). Our method combines graph semi-supervised learning, self-supervised learning, and pseudo-labeling to learn prototype representations of new classes in a zero-shot way. In contrast to existing solutions from the vision domain, POWN does not require data augmentation techniques for node classification. Experiments on benchmark datasets demonstrate the effectiveness of POWN, where it outperforms baselines by up to 20% accuracy on the small and up to 30% on the large datasets. Source code is available at https://github.com/Bobowner/POWN.

Cite this Paper

BibTeX
@InProceedings{pmlr-v274-hoffmann25a, title = {POWN: Prototypical Open-World Node Classification}, author = {Hoffmann, Marcel and Galke, Lukas and Scherp, Ansgar}, booktitle = {Proceedings of The 3rd Conference on Lifelong Learning Agents}, pages = {672--691}, year = {2025}, editor = {Lomonaco, Vincenzo and Melacci, Stefano and Tuytelaars, Tinne and Chandar, Sarath and Pascanu, Razvan}, volume = {274}, series = {Proceedings of Machine Learning Research}, month = {29 Jul--01 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v274/main/assets/hoffmann25a/hoffmann25a.pdf}, url = {https://proceedings.mlr.press/v274/hoffmann25a.html}, abstract = {We consider the problem of true open-world semi-supervised node classification, in which nodes in a graph either belong to known or new classes, with the latter not present during training. Existing methods detect and reject new classes but fail to distinguish between different new classes. We adapt existing methods and show they do not solve the problem sufficiently. We introduce a novel end-to-end approach for classification into known classes and new classes based on class prototypes, which we call Prototypical Open-World Learning for Node Classification (POWN). Our method combines graph semi-supervised learning, self-supervised learning, and pseudo-labeling to learn prototype representations of new classes in a zero-shot way. In contrast to existing solutions from the vision domain, POWN does not require data augmentation techniques for node classification. Experiments on benchmark datasets demonstrate the effectiveness of POWN, where it outperforms baselines by up to 20% accuracy on the small and up to 30% on the large datasets. Source code is available at https://github.com/Bobowner/POWN.} }
Endnote
%0 Conference Paper %T POWN: Prototypical Open-World Node Classification %A Marcel Hoffmann %A Lukas Galke %A Ansgar Scherp %B Proceedings of The 3rd Conference on Lifelong Learning Agents %C Proceedings of Machine Learning Research %D 2025 %E Vincenzo Lomonaco %E Stefano Melacci %E Tinne Tuytelaars %E Sarath Chandar %E Razvan Pascanu %F pmlr-v274-hoffmann25a %I PMLR %P 672--691 %U https://proceedings.mlr.press/v274/hoffmann25a.html %V 274 %X We consider the problem of true open-world semi-supervised node classification, in which nodes in a graph either belong to known or new classes, with the latter not present during training. Existing methods detect and reject new classes but fail to distinguish between different new classes. We adapt existing methods and show they do not solve the problem sufficiently. We introduce a novel end-to-end approach for classification into known classes and new classes based on class prototypes, which we call Prototypical Open-World Learning for Node Classification (POWN). Our method combines graph semi-supervised learning, self-supervised learning, and pseudo-labeling to learn prototype representations of new classes in a zero-shot way. In contrast to existing solutions from the vision domain, POWN does not require data augmentation techniques for node classification. Experiments on benchmark datasets demonstrate the effectiveness of POWN, where it outperforms baselines by up to 20% accuracy on the small and up to 30% on the large datasets. Source code is available at https://github.com/Bobowner/POWN.
APA
Hoffmann, M., Galke, L. & Scherp, A.. (2025). POWN: Prototypical Open-World Node Classification. Proceedings of The 3rd Conference on Lifelong Learning Agents, in Proceedings of Machine Learning Research 274:672-691 Available from https://proceedings.mlr.press/v274/hoffmann25a.html.

Related Material